Conv2dBlock#

class deeplay.blocks.conv.conv2d.Conv2dBlock(*args, **kwargs)#

Bases: BaseBlock

Convolutional block with optional normalization and activation.

Attributes Summary

expected_input_shape

Methods Summary

`call_with_dummy_data`()	Calls the forward method with dummy data to build the block.
`get_default_activation`()	Returns the default activation function for the block.
`get_default_merge`()	Returns the default merge operation for the block.
`get_default_normalization`()	Returns the default normalization function for the block.
`get_default_shortcut`()	Returns the default shortcut function for the block.
`normalized`([normalization, mode, after])
`pooled`([pool, mode, after])
`strided`(stride[, remove_pool])
`transposed`([transpose, mode, after, ...])
`upsampled`([upsample, mode, after])

Attributes Documentation

expected_input_shape#

Methods Documentation

call_with_dummy_data()#: Calls the forward method with dummy data to build the block.

get_default_activation() → DeeplayModule#: Returns the default activation function for the block.

get_default_merge() → MergeOp#: Returns the default merge operation for the block.

get_default_normalization() → DeeplayModule#: Returns the default normalization function for the block.

get_default_shortcut() → DeeplayModule#: Returns the default shortcut function for the block.

normalized(normalization: ~typing.Type[~torch.nn.modules.module.Module] | ~deeplay.module.DeeplayModule = <class 'torch.nn.modules.batchnorm.BatchNorm2d'>, mode='append', after=None) → Self#

pooled(pool: Layer = Layer[MaxPool2d](kernel_size=2, stride=2), mode='prepend', after=None) → Self#

strided(stride: int | tuple[int, ...], remove_pool=True) → Self#

transposed(transpose: Layer = Layer[ConvTranspose2d](kernel_size=2, stride=2, padding=0), mode='prepend', after=None, remove_upsample=True, remove_layer=True) → Self#

upsampled(upsample: Layer = Layer[ConvTranspose2d](kernel_size=2, stride=2, padding=0), mode='append', after=None) → Self#